Documente Academic
Documente Profesional
Documente Cultură
JUNE 1975
J. AIRCRAFT
High-Lift Aerodynamics
A. M. O. Smith
McDonnell Douglas Corporation, Long Beach, Calif.
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
Nomenclature
CL
Cn
c.
CM
C^
chord
section profile drag coefficient
local skin-friction coefficient T w /(l/2) puj
section lift coefficient
lift coefficient
conventional pressure coefficient (p p )/ (1/2)
pressure coefficient when local flow is sonic
canonical pressure coefficient (p p 0 )/(\/2)pu$
suction quantity coefficient Q/u Ox
= blowing momentum coefficient (2uft/u^ c),incompressible flow
= blowing momentum coefficient referred to momentum thickness of the boundary layer at blowing
location, uft/ujB, incompressible flow
f
H
L
m
=
=
=
=
=
M
p
q
Q
R
R6
=
=
=
=
=
=
= Stratford's separation constant (4.10); also peripheral distance around a body or wing area
= blowing slot gap, also thickness ratio of a body
= velocity in x-direction
= initial velocity at start of deceleration in canonical
and Stratford flows
= velocity normal to the wall
= a general velocity
= length in flow direction, or around surface of a body
measured from stagnation point if used in connection with boundary-layer flow
/
u
u0
v
V
x
uee/v
Presented as Paper 74-939 at the AIAA 6th Aircraft Design, Flight Test and Operations Meeting, Los Angeles, Calif., August 12-14, 1974; submitted August 14, 1974; revision received December 27, 1974.
Index categories: Aircraft Aerodynamics (including Component Aerodynamics) Boundary Layers and Convective Heat TransferTurbulent;
Subsonic and Transonic Flow.
*Note: The 36th Wright Brothers Lecture by Herman Schlichting was published in the April 1974 AIAA Journal but was incorrectly identified as
the 37th Lecture.
501
J. AIRCRAFT
A. M. O. SMITH
502
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
Greek
a
= angle of attack
6
= flap deflection
6*
= displacement thickness of the boundary layer
8
= momentum thickness of the boundary layer
v
= kinematic viscosity
p
= mass density
T
= shear stress
\l/
= stream function
Subscripts
e
= edge conditions
J
= Jet
= lower
o
= reference conditions, as in Stratford flows
u
- upper
w
= at the wall
1,
MODEL CHARACTERISTICS
<: = M.A.C.= II.02 IN.
SPAN=8.00 FT.
ASPECT RATIO = 8.69
5=7.37 FT 2
2.
Some History
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
JUNE 1975
503
HIGH-LIFT AERODYNAMICS
Figure 5 shows lift coefficient vs angle of attack for the experimental airfoil as it was modified from one to eight
elements. It generally shows that the greater the number of
elements the greater the lift; and it seems to confirm the
author's deduction, which was made three years ago, quite in
ignorance of these tests. The seven-element airfoil reached a
lift coefficient of 3.92. Tests were made at a chord Reynolds
number of about 250,000 on a wing of 6~in. chord and 36-in.
span.
Handley Page appears to have followed an empirical approach in his efforts. Concurrently Lachmann7 at Gottingen
was studying the problem theoretically. Lachmann used conformal-transformation methods and represented a slat by vor-
SXs>
~~j+ Angle
4.0
3.2
2.4
3
2 SLOTS
NO SLOT
0.8
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
J. AIRCRAFT
A. M. O. SMITH
504
10
20
30
ANGLE OF ATTACK (DEG)
40
50
Fig. 5 Ci vs ct data for the RAF 19 broken up into different numbers of elements, as indicated by number of slots.
1,504,663
O. WRIGHT ET AL
AIRPLANE
3 Sheets-Sheet 1
Fig. 6 Variable camber Albatross Biplane9 model, showing approximate range of flap angles tested.
HIGH-LIFT AERODYNAMICS
JUNE 1975
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
3.
3.1
505
A
B
C
D
E
F
G
THEORE1 ICAL
" LIMIT- 47T
WRIGHT FLYER
SPIRIT OF ST. LOLJIS
C-47
23012 AIRFOIL
B-32
C-54
C-124
12.0
<
4.0
>
SPL IT FLA PS
H 1 J K L MN -
749
1049
C-130
MA4
L-19
727
C-5A
V
K
3.0
r^ fH^
\-&
-^
J^.--
2.0
A
o _^1.0
^ s
-t
D
FOWLER OR
-SLOTTED F
MAC A AlRFOILS
0
1930
1940
1960
YEAR
(3.1)
506
J. AIRCRAFT
A, M. O. SMITH
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
and
Observe that Eq. (3.8) is nearly the same as Eq. (3.5). The circular-cylinder case is represented by f = l , and again we
(45</3<90)
(3.4)
recover c f = 4?r. Equation (3.8) shows that ce greater than
4?r can be exceeded if t is greater than 1, that is, for ellipses
that are broadside to the flow. In those cases, we are not using
Consider Eq. (3.3) with 0 = 0. We have a flat-plate airfoil.
our greatest length as the reference length. We now should use
Its maximum lift coefficient is 2?r at a = 90. A well-known
the thickness instead of the length and obtain Eq. (3.9). If Eq.
formula applying to symmetrical Joukowski airfoils that
(3.9) is used, cfmax falls continuously as t increases. Since both
accounts for thickness ratio t is
formulas give ctmgx = 4?r for a circle, we once more recover 4ir
as a limit. That is the limiting value for any single-element
(3.5)
airfoil. We know that in a uniform stream L =
pV F, where F is the circulation. Of course, any airfoil can
be mapped from a circle. But the mapping does not change the
That formula gives a value that is a few percent greater for
circulation. Hence, under the restrictions we have imposed,
common values of t.
the maximum possible lift for any kind of airfoil is 4ir.
Next, consider Eqs. (3.3) or (3.4) with /3^45. That mean
Since lift is needed near the ground perhaps more than
line is a half circle. Maximum lift occurs at a. = 45, so that
anywhere else, an interesting subsidiary question is the potential flow limit at the ground. The author knows of no simple
definitive answer. However, the Douglas Neumann program
(0=45)
(3.6)
was applied to a circle that was brought nearer and nearer to
the ground, using the image system. The stagnation point was
Finally consider Eq. (3.4). The maximum occurs when /3 =
maintained at the bottom as in Fig. 9b. For zero height off the
90 and a = 0, which gives
ground the lift coefficient extrapolates to about 4.49, which is
also the value found for the classical problem of a round
cylinder lying on a streambed. Ground, therefore, greatly
cf"max ^4ir
(3.7)
^
'
reduces the maximum possible lift. We realize full well that
ground effect often increases the lift of real wings.
The limiting mean line is a complete circle. For airfoils with
some thickness, there is a definite stagnation point at the nose
3.2 Limits of Ml, CL
and another one at the tail. Hence, that limiting mean line
brings the two stagnation points together again, and we efHigh values of CL cannot be maintained indefinitely as
fectively recover the flow about a circular cylinder, as in Fig.
speed is increased, for soon surface pressures less than ab9b.
solute zero would be indicated. Let us look into the problem
The last is an extreme mean line. The half-circle is not so exbriefly, and search especially for the limits of lift rather than
treme. In fact, modern multielement flap systems at full flap
of lift coefficient. The usual equation for lift is
deflection begin to approximate the half-circle mean line
1/2
(e.g., Fig. 29). The quantity c f = 47r/(2) in Eq. (3.6) has a
L = V2PooV2(x>CLS
(3.10)
value of nearly 9. Of course, the value of c( depends on the
length used as a reference. If arc length were used instead of
An alternate form, one that uses a different expression for
conventional chord, as above, the values would not appear to
dynamic pressure, is
be so high. In fact, for the straight line, half circle, and full
circle, the theoretical values would be 6.28, 5.65, and 4.0,
(3.11)
respectively.
The mean lines that have been considered approximate
With 7 = 1 .4, we can rewrite it as
flows about airfoils having sharp trailing edges. For them, the
Kutta condition sets the circulation to such strength that the
Lip oo = 0.7(MiC L )S
(3.12)
rear stagnation point is always at the trailing edge. Such a
flow might be called a natural flow. To round out our
Since C L is known to be a function of M , , the product
discussion, we shall mention a case in which there is no sharp
Mi C L is the quantity that is of real significance, and so we
trailing edge: the ellipse family. Assume the circulation to be
seek to make statements about its value. Observe that for a
controlled in such a way that the rear stagnation point always
given value of Mi CL the lift is now proportional to the atremains at the rear end of the jc-axis, that is, at the same point
mospheric pressure.
as in symmetric nonlifting flow (see Fig. 11). Thwaites16
A gas cannot be in tension. Hence the limiting suction
proposed airfoils that followed that principle. He used area
pressure is a perfect vacuum over the entire upper surface.
suction at the rear to eliminate separation. When our conThe limiting pressure on the lower surface is stagnation
dition for the circulation is met, the lift coefficient based on
pressure.
the length of the x-axis becomes exactly
By definition, with 7=1.4,
c(=2ir (1 + t) sin a
(3.8)
(3.13)
p
If ce were based on the length of the^-axis, it would be
~ l/2p VI
0.7p , Mi
f-^
(3.9)
REAR
STAGNATION
POINT
__ _____f
1 + 0.2MI
1 + 0.2M2
(3.14)
Before proceeding to the determination of lift limits, it is interesting to pause and consider the limits of Cp values. In
high-lift testing, maximum lift is often found to occur when
the highest local velocities reach a Mach number of 1.0. The
Cp corresponding to that condition will be called C*. If a per-
JUNE 1975
HIGH-LIFT AERODYNAMICS
507
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
(3.17)
Fig. 12 A/i Cp values as a function of Mach number for four conditions. 0.7 vacuum corresponds to M\, Cp = -1.
Table 1 Limiting values of Cp at low Mach numbers
Moo
C*(M = 1)
(^(perfect
vacuum)
Cp (0.7
vacuum)
0.10
-67
0.15
-29.3
0.20
-16.3
-143
-63.6
-35.8
-100
-44.5
-25
0.40
-3.7
0.50
-2.1
-15.9
-8.9
-5.7
-11.1
-6.3
-4.0
0.30
-7
1
0.7
-(
r , 1 + 0.2MI
1+0.2M/
1 + 0.2M* )
55
, ,,
(3.15)
M^
(3.16)
0.5
0.5
0.5
1.0
1.0
1.0
2.0
2.0
upper
00
1.55
1.5
00
1.86
1.5
oo
4.97
00
upper
1.43
1.00
0.97
1.43
1.00
0.69
1.43
1.00
M,lower
MicLlower
MiCLtotal
0
0
0
0
0
0
0
0
0.27
0.27
0.27
1.28
1.28
1.28
9.75
9.75
1.70
1.27
1.24
2.71
2.28
1.97
11.18
10.75
lvl
A. M. O. SMITH
508
-17
-15
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
4.
4.1 Introduction
This and the next section constitute the body of the paper.
In combination, they expound the basic aerodynamics of
mechanical high-lift systems. Until the 1960's, our analytical
capability was insufficient to make quantitative calculations.
Now, since the tools are here, it is time to put the whole story
together, as best we know it. That is not to say that nearly all
problems have been solved, but to a certain extent the
remaining problems amount to just irritating details.
We go into considerable detail, and often rather elementary
exposition, in the interests of making this part of the paper
complete. The approach is motivated by the many misconceptions the author-has encountered. For instance, with
respect to slotted flaps, it has been suggested that if one just
knew how to shape the airfoil, higher lift could be obtained
with a one-piece airfoil, because the leakage from bottom to
top allowed by the slots amounts to a kind of short circuit.
Properly designed multielement airfoils are better, as will be
shown. Another common statement is that a slot provides a
kind of blowing boundary-layer control by the jet of *'fresh"
air that flows through it. How can it do that? It has no more
total head than the ambient stream. Admittedly, that jet of
"fresh" air has more energy than the boundary layer, but it
only has freestream total head. At best, it may be a question
of semantics. Later we shall see that the principal effect of a
slot or slat is to reduce negative pressures, rather than to
blow.
Sections 4 and 5 are a revision and, we hope, an improvement of Ref. 20, a paper given at an AGARD meeting.
There are many papers and reports on the subject of high-lift
devices, but their emphasis usually tends to be "how to."
Ours is "why." One useful document is Ref. 21, which contains a large collection of lectures on the subject of high-lift
devices. Another is Thain's, 22 which is an extensive review of
the experimentally observed effects of Reynolds number on
high lift. It too is more concerned with overall results than
with explanation of the fundamental flow processes. Let it be
noted that our discussion in these sections is confined to twodimensional flow.
In view of Bernoulli's law, if a surface is to lift, the velocity
over the upper surface must be speeded up. But we know that
J. AIRCRAFT
for any trailing edge, even the cusped type, the flow decelerates to below freestream values. Hence, to get more lift, we
need higher velocities over the airfoil, but that in turn means
greater deceleration towards the rear. The process of
deceleration is critical, for if it is too severe, separation
develops. The science of developing high lift, therefore, has
two components: 1) analysis of the boundary layer, prediction
of separation, and determination of the kinds of flows that
are most favorable with respect to separation; and 2) analysis
of the inviscid flow about a given shape with the purpose of
finding shapes that put the least stress on the boundary layer.
The two parts amount to a kind of applied-load and
allowable-load problem. An analysis cannot be valid unless its
elements are sound, so let us first look at the problem of
predicting the onset of separation.
4.2 Accuracy of Predicting Separation Points
2.0
/?^=3.3xl06
a =10.5
1.6
1.2
0.8
0.4
PREDICTION OF SEPARATION BY
EXPERIMENT
CS METHOD
HEAD
GOLDSCHMIED
STRATFORD
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
JUNE 1975
509
HIGH-LIFT AERODYNAMICS
In almost all design work, pressure distributions are presented in terms of Cp = (p-p )/(l/2pi). In that kind of
presentation, high negative values of Cp invariably look bad,
and one is unable to tell by inspection much about the margin
of safety of the boundary layer against separation. Yet we
know from basic scaling considerations that if two pressure
distributions can be made congruent by proper scaling in the
x- and Cp-directions, then the two flows are identical except
for the Reynolds number effect, which is weak. Then, if
separation occurs, it will be at the same scaled point for both
flows. A particular 2-in. airfoil model at 100 mph will have
very high values of velocity gradients, but a similar 200-in.
model at 1 mph will have extremely low values. Yet the flows
are exactly similar because their Reynolds numbers are the
same. It is the dimensionless shape that counts. Hence it is
particularly useful to scale out the magnitude of the velocity
and also to scale out the chord. Where separation is important, the best scaling factor is the velocity just before
deceleration begins. Because all pressure distributions are put
in a standard form, it is natural to call the_m canonical
pressure distributions and use a new variable Cp. A typical
one is illustrated in Fig. 16, which shows the idea and basic
relations. The exact details of the normalization may well
depend on the problem and the nature of the pressure
distribution. The canonical pressure distribution, together
with a Reynolds number, completely describes the flow. A
meaningful Reynolds number is Re at the beginning of
pressure rise. The left-hand part in Fig. 16 might represent the
nose of an airfoil. Distance x is measured along the surface,
but the origin of x is a matter of convenience. Often it is convenient to locate it at the beginning of pressure rise, as in the
figure. Separation may occur at some point as noted. Because
of the very simple relation between pressure coefficient and
velocity ratio, the term pressure distribution is used indiscriminately for plots of either pressure coefficient or
velocity ratio, provided that the flow is at low speed.
In the canonical system C ^ ^ O represents the start of
pressure rise and Cp = +1 the maximum possible,_that is,
ue = Q. Normally there are no negative values of Cp. Furthermore, if two pressure distributions can be made congruent
by proper scaling, a flow having a deceleration of (ue/u x )2
from 20 to 10 is no more and no less likely to eparate than
one decelerating from 1.5 to 0.75 or even from 0.10 to 0.05.
The canonical plot is the one that is meaningful in separation
analysis. With magnitude effects scaled out, much more can
be told by a simple inspection than by a conventional plot. We
dwell on the canonical pressure distribution at some length,
because most working aerodynamicists do not realize its
value. Figures 17 and 18 show conventional and canonical
pressure plots for a typical high-lift airfoil. The basic charac-
100.0
10.0
(EXPERIMENTAL)
1.0
O.I
O.I
1.0
10.0
100.0
X I06 (CALCULATED)
10
20
30
_ _40_ _ 50
__
60
,jrO _
8Q/
90
j 100
JiO__-^
1.0 0
0.5
1.0 0
05
1.0 0
0.5
1.0
A. M. O. SMITH
510
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
(ue/u(K)2(x)=(ue/u0)2(x)-(u0/uOD
)2
(4.1)
J. AIRCRAFT
The unit Reynolds number is 106 per ft in Fig. 19 and 107 per
ft in Fig. 20. Pressure rise is set to start at ;c = 0, but forward
of that point are various lengths of flat-plate flow. It seemed
more convenient to construct the plots in terms of feet and
u I v rather than in terms of Reynolds number, although
conversion into Reynolds number is easy. The total region of
pressure rise is seen to be 1 ft. Four lengths of flat-plate run
were studied: 1/64 ft, 1/16 ft, 1/4 ft, and 1 ft. The initial
flows then developed boundary layers of various thickness at
the beginning of pressure rise, as indicated by values of R e in
the figures. The flat-plate flow is assumed to be entirely turbulent. If it were mixed laminar and turbulent, values of R 9 at
x = 0 would be less and different.
Calculations of the flat-plate and Cp -xm parts of the flow
were then made by the Cebeci-Smith method until separation
was reached. Lines cutting across the Cp=xm curves mark
separation points for the four lengths of flat-plate runs. The
straight-line or concave pressure rises permit the greatest
recovery before separation occurs. Also, the curves show that
the amount of recovery is sensitive to the length of flat-plate
run before the beginning of recovery. The separation loci,
while indeed functions of the length of flat-plate run, are
= I\ J}2sep I\ JV
-0.0625 -0.015625
(276)^ (108)
["-*-u
I
I
(4.4)
and
Cp = l-(ue/u0)2
(4.5)
l-Cr
(4.6)
we have
Cp(x) = l-
l-Cr
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
JUNE 1975
HIGH-LIFT AERODYNAMICS
511
l-(ue/u0)2=x
(4.8)
w+
(4.9)
If ue/u0 = 0, C^ = 1.28 at M 0 = l; and if C p = 1.0,
( w e / w 0 ) 2 = 0.18. Hence, at Cp = 1.0, the flow is still far from
stagnation. Therefore, the better canonical compressible flow
to consider is the flow in terms of (ue/u0)2. Some points
calculated thus for flow at M0- 1.0 are shown in Fig. 8 of
Ref. 20. The answer to the question of whether compressibility aggravates separation depends on the reference.
Certainly, compressibility aggravates pressure gradients on a
given bodyL which implies earlier separation. When considered in C p form, compressibility is favorable. Hence the
most fundamental approach appears to be to consider the
problem in terms of ue/u2, which is a measure of the kinetic
energy of the flow that remains. In that form, at least to M0
= 1.0, the effects of compressibility are minor when considered from the basic canonical standpoint.
In closing this section, we remark that because of_the
analytic nature of our canonical pressure distributions Cp =
x m , it is very easy to apply Stratford's criterion. That has been
done. If Stratford's predictions had been added to the plots of
Figs. 19 and 20, the differences in separation loci would appear considerable. Still they are so small that use of the charts
and the findings is not negated. Addition of Stratford-type
loci would only cause confusion. Therefore, we show only one
type of calculation, the Cebeci-Smith, which in general is
believed to be the most accurate. Furthermore, it should be
noted that Figs. 19 and 20 differ from their earlier forms in
Ref. 20. The main reason is that here we used the analytic
nature of the Cp=xm flows to compute certain necessary
derivatives.. In the earlier work, finite-difference formulas
were used. Disagreements such as those just indicate the state
of the art of turbulent-flow calculation.
4.6 Limiting Canonical Distributions
The Cp =xm families are very useful for almost all practical
flow problems, but of course the shapes of the Cp(x) curves
are arbitrary; the shapes are selected as a matter of analytical
convenience. Just as in many other problems, there is one
shape that is "best," Stratford's solution. We do not mean to
imply that the solution is exact, but as is indicated by Fig. 14,
it has acceptable accuracy. Stratford has derived a formula,
Ref. 25, for predicting the point of separation in an arbitrary
decelerating flow:
Cp[x(dCp/dx)]
6
(10 -
1/2
___
(4.10)
VALUES OF ffn
U0 'V
A-0- FT.
I0 6
1/64
108
517
1/16
276
1511
I0 7
1/4
759
4490
2196
13930
J. AIRCRAFT
A. M. O. SMITH
512
1/5
f o r C / , < ( / i - 2 ) / ( n + l)
and
C =1-
n-2
n +1
(4. lib)
recoveries may be very rapid. When the initial run is long and
the boundary layer is thick, the allowable average pressure
gradient is much less. Or conversely, thick boundary layers
are much more likely to separate than thin.
4) The unit Reynolds number effect is rather small [see
Fig. 21 and Eq. (4.1 la)].
5) Theoretically, 100% of the dynamic pressure can be
recovered, but the distance required is infinite.
6) Aside from error in the theory, the curves of Fig. 21 are
the shortest possible pressure recoveriesthey are the "end of
the line." Nothing better can be done except by boundarylayer control.
7) The Stratford pressure distribution is the path of least
resistance connecting two pressure points A and B, as will
now be shown.
If we accept quadrature formulas such as Spence's or
Truckenbrodt's, then Stratford's flow is found to be the
minimum-drag flow. Consider the flow situation sketched in
Fig. 22. We wish to go from some point A to another point B.
Various paths may be possible, as sketched. The Stratford
path is the path of lowest drag. We shall show this by means
of Truckenbrodt's formula. 26 For purposes of demonstration, we can write it as follows:
= [eA+K\*B
J
\333dx]
(4.12)
*4
Fig. 22
(4.13)
STRATFORD
Fig. 23
Two-dimensional diffuser.
HIGH-LIFT AERODYNAMICS
JUNE 1975
The width of the diffuser is w = b + 2x tan a. When that expression is combined with the equation of continuity and substituted into Eq. (4.13), we have
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
=/
~" [l+(2x/b)tana]
(4.14)
513
xlO 6 .
A. M. O. SMITH
514
J. AIRCRAFT
LAMINAR
ROOFTOP
Rc = 5xl0 6
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
=2.31
=0.0055
cdu =0.0053
cdt =0.0002
--I.OH
0.2
0.4
0.6
0.8
1.0
Fig. 27 The Liebeck airfoil that produces maximum lift with fully attached flow, design Rc = 5 x 106. Pat. Pend.
-3.0
TRANSITION
-2.0-
STRUT
-i.o
4.8 Examples
Fig. 28
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
JUNE 1975
515
HIGH-LIFT AERODYNAMICS
Until now, our concern has been with fluid flowing into
regions of higher pressure while it is in contact with a surface,
that is, a decelerating boundary-layer flow. But there can be
another kind of flow, the flow of wakes that may be out of
contact with any wall, into regions of higher pressure. Such a
flow occurs, to a degree, on any multielement airfoil. A slat
develops its own boundary layer, which flows off its trailing
edge, forming a wake of low-energy air that now flows
alongside the rest of the airfoil and on downstream. Consider
the geometry of Fig. 17, for instance, or the streamline pattern of Fig. 29. Each forward element produces wake components over its downstream partners.
The theory of that kind of wake flow is not nearly so well
developed as the theory of boundary-layer flow. Therefore,
we shall be content to given only a brief discussion of its
features, chiefly for the purpose of calling attention to them.
There can be two kinds of wakes flowing into a region of
higher pressure. One is separated from the adjacent boundary
layer by a region of potential flow. That kind occurs when
gaps are large. The other kind is so close to the adjacent boundary layer that the two flows finally merge and become one
thicker boundary layer. That kind occurs when gaps are
small. By some it is called confluent boundary-layer flow.
Because a wake is usually near the main airfoil surface, the
pressures impressed on it are little different from those on the
airfoil surface; for example, consider the streamline in Fig. 29
Fig. 29 Calculated streamline flow field for airfoil with leading edge
slat and double slotted flap, a = 0, ct = 3.70.
STATION 1
that leaves the slat. If the pressure rise is great enough, we can
have flow reversal in the stream, entirely off the surface. The
phenomenon is easily demonstrated by resorting to Bernoulli's equation. Consider a flow as illustrated in Fig. 30. At
station 0, there is a flow in which static pressure is constant
across the stream, but there is a wake-like portion in which
velocities are deficient. The wake flows into a region of higher
pressure at station 1. What happens to the wake region?
Because we are considering regions very thin with respect to
any overall curvature of the flow, it is reasonable to assume
that the static pressure p l at station 1 is also constant across
the boundary layer. Because gradients du/dy will be small, the
shear stresses will be low and then it is a good approximation
to assume that each streamline maintains its total head between station 0 and station 1. Such an assumption has been
confirmed in numerous analyses, including Stratford's.
Hence, using U for potential regions and u for energy
deficient regions, we can write, for incompressible flow,
(4.15)
and
(4.16)
l-Cn
l-Cn
l-Cn
(4.19)
J. AIRCRAFT
A. M. O. SMITH
516
0.0463'I
R"5
(4.20)
1-Cn
(4.21)
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
517
HIGH-LIFT AERODYNAMICS
JUNE 1975
is possible without causing separation. Provided that boundary-layer control is not used, our only means of obtaining
higher lift is to modify the geometry of the airfoil. Considerations that guide the modification, then, are the subject
of this section. It is helpful to think in terms analogous to
those of structural designapplied and allowable loads. The
airfoil applies the loads, and the boundary layer determines
the allowable. Separation of a boundary layer may be likened
to reaching the yield point in materials testing. Initial
separation rarely coincides with the maximum lift that an airfoil develops, for the lift usually continues to increase.
Likewise, the yield point in a metal is not the point of ultimate
load. That usually is the rupture point. Hence, the point of
maximum lift coefficient may be likened crudely to the rupture point of a material. The analogy is quite rough, of
course, but it is mentioned because the interaction between
boundary layer and shaping of an airfoil is not very widely appreciated. We close these introductory remarks by observing
that aerodynamic science has advanced to the point where we
can satisfactorily predict the point of initial separation (the
yield point) but not the condition of maximum lift (the rupture point). The second problem is still beyond the state of the
art, but it is assuming high priority because of successful
solution of the simpler problems.
5.2 Single-Element Airfoils
- DEFLECTED 25*
-VARIABLE CAMBER
-2.0
-1.6
-1.2
TEST DATA
- THEORY
-0.8
-0.4
0.4
0.8
20
40
60
PERCENT CHORD
80
100
LOWER
0.2 -
20
40
60
80
100
PERCENT CHORD
Fig. 32 Comparison of two kinds of flaps on a NACA 63A010 airfoil. For the plain flap a = 0 and inviscid ct = 1.78. The airfoil with
variable camber flap was set at a = 1.06 in order to obtain the same
Cg. Separation points are marked by arrows. Rc = 107. Transition is at
forward suction peak.
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
518
A. M. O. SMITH
J. AIRCRAFT
o.o
a-15
cj
OF ISOLATED
AIRFOIL = 2.374I
eg TOTAL
c^ AIRFOIL
c , AIRFOIL
= 2.4741
= 1.8789
= 0.3680
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
JUNE 1975
519
HIGH-LIFT AERODYNAMICS
A 18.00
jj
AIRFOIL =0.98
\ C ON 3
AIRFOIL =0.49
TOTAL SYSTEM = 1.54
30
40
50
PERCENT CHORD
60
2.0
2.2
2.4
2.6
2.8
3.0
A. M. O. SMITH
520
a=(5
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
cj OF ISOLATED
AIRFOIL = 2.5937
c/ TOTAL = 3.4937
Cj AIRFOIL = 3.3631
C. AIRFOIL = -0.0763
Circulation Effect
J. AIRCRAFT
,0=0.50
Q.25C
'/D-0.5C
=0.25C
2.0 -
521
HIGH-LIFT AERODYNAMICS
JUNE 1975
0 SOLUTION
CIRCULATION SOLUTION
90 SOLUTION
^a = \5 ISOLATED AIRFOIL
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
D-Q.25C
= 10 ISOLATED AIRFOIL
0.25<r
D=0.5c
= 5 ISOLATED AIRFOIL
30
60
S DEGREES
CONFIGURATION
ISOLATED AIRFOIL
0.25C CIRCLE
0.50C CIRCLE
O.I
0.2
0.3
12.8642
6.5950
4.O929 0
0.4
0.5
20.2548
13.3444
14.0289
0.6
0.8800
1.1472
1.1855
0.7
0.8
6, and
0.043
0.086
0.9
1.0
The velocity ue on the surface of an isolated circular cylinder in crossflow is ue/u ^ = 2 sin j3, where /3 is angle from the
nose. For our problem, the velocity component that is normal
to the airfoil is 2 sin 0 cos /3 = sin 2/3. The maximum value of
that component, which occurs at 0 = 45, is not much less
than the 6 = 60 maximum seen in Fig. 41. The rough coincidence justifies the assumption that the change in circulation
is a function of the cross flow induced at the trailing edge by
other bodies.
Another aid in thinking about the problem of modifying the
circulation is to consider the rear stagnation point in nonlifting flow. Without circulation, the rear stagnation point is on
the upper surface at some distance forward of the trailing
edge. The farther it is moved forward, by whatever means, the
greater the circulation required to move it back to the trailing
edge. The two guidelinescross-flow strength and nonlifting
stagnation-point location are the best that can be proposed as
means of understanding and designing to maximize the cir-
A. M. O. SMITH
522
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
5.6
Dumping Effect
Closely related to the circulation effect is the dumping effect. The favorable interference of a downstream element induces cross flow at the trailing edge that enhances circulation
of the upstream element. But the interference may also increase velocities in a tangential direction so that the flow from
a forward element is discharged into a higher velocity region,
thus reducing pressure-recovery demands. The effect is quite
favorable to the boundary layer. According to Eq. (4.4), the
suction lift can be increased in proportion to (u T E ) 2 for the
same margins against separation.
Does the effect really exist? Yes, indeed! It can be seen in
any properly designed multielement airfoil. It is clearly shown
theoretically in Fig. 42; values of the dumping-velocity ratio
are tabulated in the right-hand columns of the table in the
figure. For the three casesplain airfoil, airfoil plus 0.25c circle, and airfoil plus 0.50c circlethe velocity ratios are 0.88,
1.147, and 1.186, respectively. The canonical ratio nearly
doubles; it increases from 0.043 to 0.085.
Figure 36 is a theoretical case for an airfoil with slat. For
the main airfoil, the dumping-velocity ratio squared is 0.85.
On the slat, it is much higher: 2.35. The pressure distributions
are also shown in canonical form, and according to them the
slat is less severely loaded than the main airfoilin the sense
of margin against boundary-layer separation.
Figure 44 shows experimental data for a three-element airfoil. Again the dumping-velocity effect is clearly displayed,
this time confirmed by experiment. For the three surfaces,
starting with the flap, the dumping-velocity-squared ratios
(ue/u oo ) 2 are 0.67, 2.0, and 2.28. The canonical plots indicate that, for the conditions shown, the main airfoil is less
severely loaded than the slat and the flap. The enhanced
dumping velocity shown is typical. The pressure distribution
of any properly designed multielement airfoil shows it; one
only need look for it.
It is interesting to conjecture what might be done with the
effect if an inverse design method for multielement airfoils
became available, assuming that the complete airfoil
requirements permitted. Start with some desired pressure
distribution for a hypothetical three-element airfoil, for
J. AIRCRAFT
(Ct/c{o)=f!+f2m2+f3(m2m3)
(5.2)
JUNE 1975
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
5.7
523
HIGH-LIFT AERODYNAMICS
On each element of a properly designed multielement airfoil, the boundary layer starts out afresh. It is well known, of
course, that thinner boundary layers can sustain greater
pressure gradients than thicker ones. That fact is demonstrated well by Figs. 19 and 20. It can be displayed in a more
analytic form by Stratford's formula, Eq. (4.10), which we
here repeat.
] V2=(10-6R)1/WS
(4.10)
BASE
DOMPING
VELOCITY
x/c
524
A. M. O. SMITH
J. AIRCRAFT
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
Cp
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
6.
6.1
525
HIGH-LIFT AERODYNAMICS
JUNE 1975
Power-Augmented Lift
General
A few remarks should be made about the subject of poweraugmented lift. Power properly applied, as through boundary-layer control, can greatly delay separation and hence increase maximum lift. Powered lift augmentation usually
requires pumps and internal ducting, which amountsat
leastto weight and cost complications. There is no doubt
that higher lifts can be obtained, but whether they make for a
better total airplane system is a moot question. The answer
depends on the efficiency of the lift-augmentation system.
Here we shall confine ourselves to three studies that touch on
the theoretical performance possibilities, together with ideas
for improving the overall system.
6.2 Area Suction
The thorough studies reported in Ref. 44 show that turbulent flows with mass transfer can be calculated with nearly
2.0
1.0
I
FRONT
4
5
ELEMENT NO.
8
REAR
Fig. 48 Upper surface dumping velocity values for the airfoil and
conditions of Fig. 4. The ratio ( W T E / w m a x ) 2 for the ensemble is
0.0045.
a = 10'
526
A. M. O. SMITH
J. AIRCRAFT
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
(6.2)
Qc
cQc- -^~
unx
i (y u
= \
dy
x J o Un
(6.4)
(6.5)
CQc = (7/8)(d/x)(y/d)8c"
(6.6)
From Eq. (6.3) and the first of Eq. (6.5) we can also write
When this, together with the second of Eq. (6.5) are substituted into (6.6) we obtain
CQc= [0.324 / (u
(6.7)
This formula shows that if the demanded pressure rise is indeed low the required suction C Q is very low. If u Ox/v is 107
then CQc = 0.013 C*f. If Cpc = 0.c5, the required CQc is only
0.0008 but for reaching stagnation pressure it is 0.013. Unlike
the results on Figs. 50 and 51 some suction is indicated for
even the smallest pressure rise, but the fourth power variation
shows the demand to be exceedingly small. The most interesting result from this analysis is the fourth power relation
ofCQtoCp.
In Lachmann, 7 Wuest presents theoretical studies made by
Pechau's approximate method. The studies were of area suction applied to simple airfoils in various ways. Suction
requirements for avoiding separation are given. Generally,
our calculations show the same C Q = Q/u Ox magnitude as his,
that is, of 0(10-3). We have not presented specific examples,
because charts like Figs. 50 and 51 are more general. For a
real engineering problem, since the boundary-layer computing
method is entirely general, the particular case should be and
can be studied carefully. Charts are then just a guide for
preliminary estimation purposes.
The results shown represent no kind of optimum. The start
of suction was arbitrarily chosen, and the intensity chosen was
HIGH-LIFT AERODYNAMICS
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
JUNE 1975
527
(6.8)
pujO
ujt
ujB
(6.9)
C=2(uj/ul)(e/c)Cit
(6.10)
A. M. O. SMITH
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
528
AIRFOIL 1,
J. AIRCRAFT
show the specific steps that are taken in finding a shape. The
steps, slightly paraphrased, are as follows:
1) Specify the desired airfoil velocity distribution and jet
momentum coefficient.
2) Divide the velocity distribution into symmetric and antisymmetric components.
3) Determine the approximate jet-entrainment effects.
Since the operation is noniterative, the sink distribution representing the jet is placed on an extension of the airfoil chord,
and its strength is determined by assuming that the velocity
just outside the jet is equal to the freestrearn value. The entrainment effects at the true airfoil surface are assumed to be
the same as those calculated at the airfoil chord line.
4) Use an inverse method to find the thickness distribution
that corresponds to the symmetrical component of the
velocity specified in step 2 that remains after entrainment effects have been subtracted out.
5) Design the mean line needed to produce the antisymmetric portion of the velocity distribution, including the
influence of the jet. For that purpose, a linearized method is
used. This part of the design procedure is the only part that is
fundamentally different from conventional design methods.
For more details, see Ref. 48. Either: a) the vorticity at the
trailing edge can be specified, find the jet angle; or b) the jet
angle can be specified, find the vorticity at the trailing edge.
6) Combine camber and thickness distributions.
7) Using the general nonlinear direct jet-flap airfoil
method, check the result. If necessary, repeat the cycle with
pressures slightly modified to correct errors revealed by the
direct-method check.
We present two figures to show typical results. Figure 55
shows the effect of blowing. High blowing rates relieve the
camber, assuming that the velocity distribution is maintained
constant. The points are checks of the computed inverse shape
by means of the direct, nonlinear method using it as input
data. The difference in c( noted for the three airfoils is due to
the lift component of the jet. Each airfoil is shown at the
angle required to develop the design pressure distribution.
An extreme case, one that taxes the linearized camber treatment, is shown in Fig. 56. Here entrainment effects have been
included in the inverse method. The entrainment effects, as
well as the entire camber-shape solution, are worked out
2.5
o o o o o o o o o o o
0.2
X/C
7.
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
529
HIGH-LIFT AERODYNAMICS
JUNE 1975
The Future
References
!
Downloaded by INST TEC DE AERONAUTICA (ITA) on November 28, 2014 | http://arc.aiaa.org | DOI: 10.2514/3.59830
530
A. M. O. SMITH
J. AIRCRAFT
"Mathematical Model for Two-Dimensional Multi-Component Airfoils in Viscous Flow," NASA Contractor Rept. CR-1843, July 1971,
NASA.
31
Stevens, W. A., Goradia, S. H., and Braden, J. A.,
"Mathematical Model for Two-Dimensional Multi-Component Airfoils in Viscous Flow," AIAA Paper 72-2, San Diego, Calif., 1972.
32
Foster, D. N., Ashill, P. R., and Williams, B. R., "The Nature,
Development and Effect of the Viscous Flow Around an Airfoil with
High-Lift Devices," Tech. Rept. 72227, Jan. 1973, Royal Aircraft
Establishment, Farnborough, England.
33
Prandtl, L. and Tietjens, O. G., Applied Hydro- and
Aeromechanics, Dover, New York, p. 155.
34
Abbott, I. H., and von Doenhoff, A. E., Theory of Wing Sections, Dover, New York, 1958, p. 227.
35
Perkins, C. D. and Hage, R. E., Airplane Performance, Stability, and Control, Wiley, New York, 1954, p. 78.
36
Chen, A. W., "The Determination of the Geometries of
Multiple-Element Airfoils Optimized for Maximum Lift Coefficient,"
TM-X-67591, 1971, NASA.
37
Liebeck, R. H. and Smyth, D. N., "Study of Slat-Airfoil Combinations Using Computer Graphics," Journal of Aircraft , Vol. 10,
April 1973, p. 254-256.
38
Hess, J. L. and Smith, A. M. O., "Calculation of Potential Flow
about Arbitrary Bodies," Progress in Aeronautical Science, Vol. 8,
Pergamon Press, New York, 1966.
39
Callaghan, J. G. and Beatty, T. D., "A Theoretical Method for
the Analysis and Design of Multielement Airfoils," Journal of Aircraft, Vol. 9, Dec. 1972, p. 844-848.
^Gentry, A. E., "How Sails Work," SAIL Magazine (in several
issues between April 1973 and Oct. 1973).
41
Ashill, P. R., "A Study of the Effect of the Wake of the Main
Aerofoil of a Fowler Flap Configuration on the Lift of the Flap,"
Tech. Rept. TR 72081, July 1972, Royal Aircraft Establishment,
Farnborough, England.
42
Moser, A. and Shollenberger, C. A., "Inviscid Wake-Airfoil Interaction on Multielement High Lift Systems," Journal of Aircraft,
Vol. 10, Dec. 1972, p. 765-770.
43
Wilkinson, D. H., "A Numerical Solution of the Analysis and
Design Problems for the Flow Past One or More Aerofoils or
Cascades," R&M 3545, 1967, Aeronautical Research Council,
London.
^Cebeci, T. and Smith, A. M. O., Analysis of Turbulent Boundary
Layers, Academic Press, New York, 1974.
45
Cebeci, T. and Witherspoon, G. F., "Theoretical Suction and
Pressure Distribution Bounds for Flow Separation in Retarded
Flow," Journal of Aircraft, Vol. 11, Jan. 1974, pp. 61 -64.
^Gartshore, I. S. and Newman, B. G., "The Turbulent Wall Jet in
an Arbitrary Pressure Gradient," The Aeronautical Quarterly, Feb.
1969. p. 25.
47
Shen, C. C., Lopez, M. L., and Wasson, N. F., "A Jet-Wing
Lifting-Surface Theory Using Elementary Vortex Distributions,"
Journal of Aircraft, Vol. 12, May 1975, pp. 448-456.
48
Halsey, N. D., "Methods for the Design and Analysis of JetFlapped Airfoils," Journal of Aircraft, Sept. 1974, pp. 540-546.
49
Nash, J. F. and Patel, V. C., Three-Dimensional Turbulent
Boundary Layers, S. B. C. Tech. Books, Atlanta, Ga., 1972.
50
Cebeci, T., "A General Method for Calculating ThreeDimensional Incompressible Laminar and Turbulent Boundary
Layers; II. Three-Dimensional Flows in Cartesian Coordinates,"
Rept. MDC J6517, Mar. 1974, Douglas Aircraft Co., Long Beach,
Calif.
51
Kline, S. J., Morkovin, M. V., Sovran, G., and Cockrell, D. S.,
"Computation of Turbulent Boundary Layers," AFOSR-IFPStanford Conference, Stanford Univ. Press, 1968.
52
Cornish, J. J., Ill, "High-Lift Applications of Spanwise
Blowing," Paper No. 70-09, Seventh Congress of the ICAS, Rome,
1970.
53
Krall, E. M., and Haight, C. H., "Wind Tunnel Tests of a Trapped-Vortex High-Lift Airfoil," Rept. B-94300/3TR-10, Advanced
Technology Center, Dallas, Texas.